Erkie Asmare 1 and Andualem Begashaw

Open Acc Biostat Bioinform
Copyright © 
Erkie Asmare Beyene
6/7
How to cite this article:
Erkie A, Andualem B. Review on Parametric and Nonparametric Methods of Efficiency Analysis. Open Acc Biostat Bioinform
. 2(2). 
OABB.000534. 2018. DOI:
10.31031/OABB.2018.02.000534
Volume 2 - Issue - 2
put DEA uses linear programming to construct a non-parametric 
piece-wise surface (or frontier) over the data, so as to be able to 
calculate efficiencies without parameterizing the technology [16]. 
Therefore; DEA calculations are designed to maximize the relative 
efficiency score of each unit, subject to the constraint that the set of 
weights obtained in this manner for each DMU must also be feasible 
for all the others included in the sample [4].
The x-axis measures the input quantity and the y-axis the out-
put quantity. Each point represents the input-output combination 
of each producer. DEA will envelop all these points in order to 
compute a piece-wise frontier over them, then the efficiency score 
of each producer depends on the distance from the frontier. DEA 
constructs a piecewise linear convex frontier connecting the set of 
best practice observations. It envelopes input and output data, rel
-
ative to which costs are minimized or profit/revenue is maximized. 
Efficiency scores are then calculated from the frontiers generated 
by a sequence of linear programs. These fractional programs are 
defined by external optimization of the ratio of weighted sum of 
outputs to weighted multiple input, subject to the constraints of 
non-decreasing weights and efficiency measure less than or equal 
to one [14]. Efficiency measures can be calculated relatively to the 
efficient technology, represented by a form of frontier function. 
Then, inefficiency is the distance of the other observations from 
this best-practice realization [11].